The objective of this two-year initiative is to develop the basis for a fundamentally new technology for an in situ clinical diagnosis of solid tumors with the expectation of rapid and accurate detection and diagnosis of cancer. The research approach is interdisciplinary (acoustic and electrical engineers, image and signal processor, material scientists, and board-certified veterinary pathologist). Given the important in patient health-care management of obtaining a highly accurate and timely diagnosis of a tumor and the difficulties, processing time, and risks associated with surgical and aspirational biopsies, a technique which would permit the evaluation of tumors in situ would be enormously beneficial medically and cost effective. The specific aims of this two-year initiative are: (1) to identify ultrasonic image features of tissues and cells that differentiate cytoarchitectural features of normal tissues from selected neoplastic tissues using well-established non-invasive pulse-echo data acquisition schemes and high-definition image formation techniques applicable for the invasive in vivo ultrasonic microprobe geometry, (3) to construct and evaluate invasive in vivo ultrasonic microprobes using two separate construction approaches, and (4) to identify ultrasonic image features of tissues and cells that differentiate cytoarchitectural features of normal tissues from selected neoplastic tissues using the invasive in vivo ultrasonic microprobes. Light microscopic histopathologic evaluation of the identical tissues by a board-certified veterinary pathologist will be used as the gold standard. Three categories of ultrasonic data acquisition are proposed, viz., non- invasive using commercial transducers; invasive microprobe using existing, commercially available materials; and invasive microprobe using newly developed materials. Conventional B-mode and C-mode imaging techniques will be investigated. In addition, it is proposed to adapt signal processing techniques from synthetic aperture radar (SAR) imaging techniques. Also, image formation and enhancement methods will be tested and verified on stimulate data based on a full-fledged statistical model of the ultrasound imaging processing, adopting an enhanced version of the standard fully-developed-speckle model.